Side-by-side comparison of stars, features, and trends
WeKnora is an intelligent knowledge management framework that leverages LLMs to provide both rapid RAG-based Q&A and complex ReACT-based reasoning. The platform supports diverse data sources, multiple document formats, and seamless integration with various IM channels and LLM providers. Its modular architecture ensures full data sovereignty through local or private cloud deployment options.
FlashMLA is a library of high-performance attention kernels developed by DeepSeek to power their V3 and V3.2-Exp models. The repository provides specialized implementations for both sparse and dense attention, supporting efficient prefill and decoding stages. These kernels are designed for modern GPU architectures to deliver significant performance improvements in compute-bound workloads.